Chain type composite multi-ring resonant microwave planar micro-plastic concentration detection sensor and detection method thereof

Abstract

The application discloses a chain type composite multi-ring resonant microwave planar micro-plastic concentration detection sensor and a detection method thereof. The composite circular ring chain in the application is located in a sample loading area and comprises a plurality of composite circular ring units. Each composite circular ring unit comprises an omega type resonator structure and two open loop resonator structures arranged in the omega type resonator structure. The omega type resonator structure is used for concentrating surface current and microwave magnetic field. The openings of the two open loop resonator structures and the adjacent metal edges form a local capacitive coupling area. During detection, the sample to be detected is placed in the sample loading area, the sensor scattering parameter response is collected, microwave characteristic parameters such as resonance frequency, amplitude attenuation and phase size are extracted, the corresponding relationship between the micro-plastic concentration and the microwave characteristic parameters is established, and then the micro-plastic concentration detection result is obtained. The application can realize high-precision detection of the micro-plastic concentration in water through the composite circular ring chain structure to form a local enhanced electric field and multi-resonance response.

Description

Technical Field

[0001] This invention belongs to the field of water environment pollution detection technology, specifically relating to a chain-type composite multi-ring resonant microwave planar microplastic concentration detection sensor and its detection method. Background Technology

[0002] Microplastics refer to small plastic particles or fragments, mainly originating from the wear, aging, and degradation of plastic products, as well as industrial emissions. They are widely distributed in rivers, lakes, oceans, and wastewater treatment systems. Due to their small particle size, large specific surface area, and strong hydrophobicity, microplastics easily adsorb persistent organic pollutants and heavy metals, and migrate and accumulate through the food chain, posing a potential threat to the ecological environment and human health. Therefore, developing a technology for rapid, accurate, and low-cost detection of microplastics in water bodies is of great significance.

[0003] Currently, microplastic detection in water bodies typically employs methods such as visual observation under a microscope, scanning electron microscopy, Fourier transform infrared spectroscopy, Raman spectroscopy, and X-ray photoelectron spectroscopy. While these methods offer advantages in morphological analysis and material identification for microplastics, they generally suffer from drawbacks such as long sample testing cycles, expensive equipment, large detection errors, difficulty in in-situ detection, and low automation, making it challenging to meet the needs of rapid monitoring and on-site screening of large-scale aquatic environments.

[0004] Compared with traditional detection methods, microwave detection methods offer advantages such as fast response speed, small sensor structure size, ease of integration, and lower manufacturing cost. Microwave-based methods for detecting microplastics in water primarily utilize the difference in dielectric constant between water (with a dielectric constant of approximately 80) and microplastic contaminants (with a dielectric constant of approximately 3) to measure concentration. Since the dielectric constant of water is significantly higher than that of common plastic materials, changes in the microplastic content alter the equivalent dielectric constant and other electromagnetic parameters of the sample, thus affecting the coupling state between the microwave sensor and the sample. This causes changes in the sensor's scattering parameter response, manifested as resonant frequency shifts, amplitude changes, and phase changes. By establishing a correspondence between microplastic concentration and microwave characteristic parameters, the concentration of microplastics can be detected.

[0005] Microwave resonant ring structures can generate a significant resonant response at specific frequencies, and the electric field distribution in their sensitive region is highly sensitive to changes in the dielectric properties of the sample under test. When the sample is placed in the sensitive region of the resonant ring sensor, changes in the sample's dielectric constant will disturb the sensor's equivalent electromagnetic parameters, thereby causing changes in the resonant response. Therefore, the microplastic detection method based on microwave resonant rings can identify and characterize different microplastic concentrations by analyzing characteristic parameters such as resonant frequency, amplitude, and phase, providing a technical basis for the structural design of microplastic concentration detection sensors and the optimization of detection methods.

[0006] When using microwave sensing structures to detect microplastic concentration in water, the coupling degree between the sample loading region and the resonant sensitive region, the local electric field strength, and the number of extractable microwave characteristic parameters directly affect the detection sensitivity and inversion stability. Therefore, it is necessary to design a planar microwave sensor that can generate a localized enhanced electric field in the sample loading region, possess multiple resonant responses, and can jointly utilize characteristic parameters such as frequency shift, amplitude, and phase to characterize changes in microplastic concentration. Summary of the Invention

[0007] The purpose of this invention is to provide a chain-type composite multi-ring resonant microwave planar microplastic concentration detection sensor and its detection method. By introducing a composite circular ring chain structure into the sample loading region, a local enhanced electric field and multi-resonant response are formed, and multi-dimensional microwave characteristic parameters such as resonant frequency, amplitude and phase are extracted, thereby improving the sensor's response capability to changes in the dielectric parameters of the sample under test and the reliability of microplastic concentration detection.

[0008] To achieve the above objectives, the present invention adopts the following technical solution.

[0009] A chain-type composite multi-ring resonant microwave planar microplastic concentration detection sensor of the present invention includes:

[0010] Dielectric substrate;

[0011] The first microstrip transmission line and the second microstrip transmission line are disposed on the upper surface of the dielectric substrate;

[0012] A first rectangular patch and a second rectangular patch are connected and coupled to each other; a first microstrip transmission line is connected and coupled to the first rectangular patch; a sample loading region is formed between the first rectangular patch and the second rectangular patch; and...

[0013] A composite circular ring chain is disposed between the first rectangular patch and the second rectangular patch and located within the sample loading area; the composite circular ring chain includes multiple composite circular ring units arranged in a chain along the central axis, each composite circular ring unit including an outermost omega-type resonator structure and two open-ring resonator structures disposed inside the omega-type resonator structure.

[0014] The radii of the two open-ring resonator structures are both smaller than the radius of the omega-type resonator structure, and the opening directions of the two open-ring resonator structures are opposite. The omega-type resonator structure is used to concentrate surface current and microwave magnetic field. The openings of the two open-ring resonator structures and adjacent metal edges form a local capacitive coupling region, which is used to form a local strong electric field in the sample loading region.

[0015] A method for detecting microplastic concentration using the aforementioned chain-type composite multi-ring microwave microplastic concentration detection sensor, according to the present invention, includes the following steps:

[0016] (1) Place the sample to be tested in the sample loading area so that the sample to be tested is above the composite circular chain;

[0017] (2) The scattering parameter response of the sensor is acquired within a preset frequency range using a vector network analyzer or a microwave signal acquisition module;

[0018] (3) Extract the resonant frequency, amplitude attenuation and phase magnitude of multiple resonant frequency points from the scattering parameter response as microwave characteristic parameters;

[0019] (4) Establish the correspondence between microplastic concentration and microwave characteristic parameters based on known microplastic concentration samples;

[0020] (5) Test the sample with unknown microplastic concentration and obtain the microplastic concentration test result according to the corresponding relationship.

[0021] Compared with the prior art, the present invention has at least the following beneficial effects.

[0022] First, the present invention constructs a resonant sensitive region by setting a composite circular chain between the left and right rectangular patches, and loads the sample to be tested into this region, so that the change in equivalent dielectric parameter caused by the change in sample concentration can more fully act on the resonant structure, thereby causing a distinguishable change in the scattering parameter response of the sensor.

[0023] Secondly, the omega-type resonator structure of the present invention can concentrate current and microwave magnetic field. The opening of the open ring resonator and the adjacent metal edge can form a local capacitive coupling region. The combination of the two is conducive to forming a strong local electric field in the sample loading area, thereby improving the ability to sense changes in the concentration of microplastic samples.

[0024] Third, the chain-ring structure of the present invention can generate multiple resonance responses. Different resonance frequencies correspond to different electric field distributions and coupling modes. It can simultaneously extract multiple microwave features such as resonance frequency, amplitude attenuation and phase magnitude, providing multi-dimensional feature inputs for establishing a microplastic concentration inversion model.

[0025] Fourth, this invention can determine the main detection frequency and auxiliary detection frequency based on the scattering parameter response curve and the electric field distribution at the corresponding frequency point, and improve the fitting accuracy of the microplastic concentration inversion results by multi-frequency point and multi-parameter joint modeling, and reduce the prediction error caused by single frequency point features.

[0026] Fifth, the present invention adopts a planar microwave structure, which is simple in structure, easy to process and integrate, and is suitable for low-cost, convenient and on-site water microplastic concentration detection equipment. Attached Figure Description

[0027] Figure 1(a) is a schematic diagram of the front structure of the chain-type composite multi-ring microwave microplastic concentration detection sensor provided in an embodiment of the present invention;

[0028] Figure 1(b) is a schematic diagram of the grounding metal layer structure on the back of the sensor;

[0029] Figure 1(c) is a partially enlarged schematic diagram of the composite ring unit;

[0030] Figure 2 A schematic diagram of the detection structure when the microfluidic chip is disposed above the sample loading area, as provided in an embodiment of the present invention;

[0031] Figure 3(a) shows the sensor S under different microplastic concentration sample loading conditions provided in the embodiment of the present invention. 11 Amplitude response curve;

[0032] Figure 3(b) shows the sensor S under different microplastic concentration sample loading conditions provided in the embodiment of the present invention. 11 Phase response curve;

[0033] Figure 3(c) shows the sensor S under different microplastic concentration sample loading conditions provided in the embodiment of the present invention. 21 Amplitude response curve;

[0034] Figure 4(a) shows the S of the sensor provided in the embodiment of the present invention under sample loading conditions. 11 A schematic diagram of the electric field distribution at the 2.41 GHz resonant frequency.

[0035] Figure 4(b) shows the S of the sensor provided in the embodiment of the present invention under sample loading conditions. 11 A schematic diagram of the electric field distribution at the 2.59 GHz resonant frequency.

[0036] Figure 4(c) shows the S of the sensor provided in the embodiment of the present invention under sample loading conditions. 11 A schematic diagram of the electric field distribution at the 2.92 GHz resonant frequency.

[0037] Figure 4(d) shows the S of the sensor provided in the embodiment of the present invention under sample loading conditions. 11 A schematic diagram of the electric field distribution at the 3.56 GHz resonant frequency.

[0038] Figure 5(a) shows the low-frequency S 11 The model's prediction results for microplastic concentration;

[0039] Figure 5(b) shows the high-frequency S 11 The model's prediction results for microplastic concentration;

[0040] Figure 5(c) shows the S-based 21 The model's prediction results for microplastic concentration;

[0041] Figure 5(d) shows the microplastic concentration prediction results based on the fusion model;

[0042] Figure 6 The absolute error of microplastic concentration based on the fusion model is provided for the embodiments of the present invention.

[0043] Explanation of reference numerals in the attached figures: 1. Dielectric substrate; 2. First microstrip transmission line; 3. Second microstrip transmission line; 4. First rectangular patch; 5. Second rectangular patch; 6. Composite circular ring chain; 7. Sample loading area; 8. Sample under test; 9. First port; 10. Second port; 11. Composite circular ring unit; 12. Omega-type resonator structure; 13. Open-loop resonator structure; 14. Open-loop resonator opening; 15. Localized capacitive coupling region. Detailed Implementation

[0044] The present invention will be further described below with reference to the accompanying drawings and embodiments. The following embodiments are for illustrative purposes only and are not intended to limit the scope of protection of the present invention. Unless otherwise specified, the embodiments and technical features described in the present invention can be combined with each other.

[0045] This application includes a dielectric substrate, a first microstrip transmission line disposed on the dielectric substrate, a second microstrip transmission line, a first rectangular patch, a second rectangular patch, and a composite circular chain.

[0046] A first microstrip transmission line is connected to and coupled to a first rectangular patch, and a second microstrip transmission line is connected to and coupled to a second rectangular patch. A sample loading region is formed between the first and second rectangular patches. A composite circular ring chain is disposed between the first and second rectangular patches and located within the sample loading region.

[0047] The composite circular ring chain comprises multiple composite circular ring units arranged in a chain along a central axis. There are four composite circular ring units, each including an outermost omega-type resonator structure and two open-ring resonator structures disposed within the omega-type resonator structure. The radii of the two open-ring resonator structures are both smaller than the radius of the omega-type resonator structure, and their opening directions are opposite. The omega-type resonator structure is used to concentrate surface current and microwave magnetic field; the openings of the two open-ring resonator structures and their adjacent metal edges form a local capacitive coupling region, used to generate a strong local electric field within the sample loading region, thereby improving the sensor's response sensitivity to changes in the dielectric constant of the sample under test. This sensor employs a two-port structure, facilitating the analysis of the sensor's scattering parameter response, electromagnetic coupling effect, electric field distribution, and resonant frequency. The sample under test is placed between two rectangular patches and covers the sensitive area above the composite circular ring chain.

[0048] This application also provides a method for detecting microplastic concentration using the above-mentioned sensor, comprising: placing the sample to be tested in the sample loading area; acquiring the scattering parameter response of the sensor through a vector network analyzer or a microwave signal acquisition module; extracting microwave characteristic parameters such as resonant frequency, amplitude attenuation, phase magnitude, and resonant offset from the scattering parameter response; establishing the correspondence between microplastic concentration and multiple microwave characteristic parameters; and obtaining the microplastic concentration detection result based on the microwave characteristic parameters of the sample to be tested.

[0049] In a specific simulation example, the concentration of the microplastic suspension sample was set to 63.60 mg / mL, and the frequency sweep range was from 2 GHz to 4.5 GHz. The scattering parameter response curves of the sensor and the electric field distribution at the corresponding resonant frequencies were obtained. Simulation results show that under this sample loading condition, the sensor exhibits different degrees of resonant response near 2.41 GHz, 2.59 GHz, 2.92 GHz, and 3.56 GHz. Among them, the S at 3.56 GHz... 11 The amplitude attenuation is greatest at 2.92 GHz, and the overall electric field is strong in the circular chain, rectangular patch, and microstrip transmission line regions, making it the primary detection frequency in this simulation example. At 2.92 GHz, the electric field distribution in the circular chain region is also strong, and the resonant unit is more susceptible to changes in dielectric constant, making it the auxiliary detection frequency in this simulation example. When the concentration of microplastics in the sample decreases, the equivalent dielectric constant of the sample loading region increases, and the equivalent capacitance of the sensor increases. According to the LC resonance relationship, an increase in equivalent capacitance leads to a decrease in the resonant frequency; therefore, the main resonant frequency of the sensor shifts to lower frequencies. In addition, the S values ​​on both sides of the resonant frequency point... 11 The amplitudes also exhibit monotonically varying ranges. By utilizing the resonant frequency offset and the phase change of the resonant amplitude, an inversion model for the microplastic concentration of the sample under test can be established.

[0050] Example:

[0051] As shown in Figures 1(a), 1(b), and 1(c), this embodiment provides a chain-type composite multi-ring microwave microplastic concentration detection sensor. The sensor includes a dielectric substrate 1, a first microstrip transmission line 2, a second microstrip transmission line 3, a first rectangular patch 4, a second rectangular patch 5, and a composite circular ring chain 6.

[0052] The first microstrip transmission line 2 and the second microstrip transmission line 3 are respectively disposed on the upper surface of the dielectric substrate 1 and are used to transmit microwave excitation signals. The first port 9 and the second port 10 are respectively disposed on the side of the dielectric substrate 1 and are connected to the first microstrip transmission line 2 and the second microstrip transmission line 3. The first port 9 and the second port 10 are used to input and output microwave excitation signals and connect to a vector network analyzer or a microwave signal acquisition module.

[0053] The first microstrip transmission line 2 is connected to and coupled to the first rectangular patch 4, and the second microstrip transmission line 3 is connected to and coupled to the second rectangular patch 5. A sample loading region 7 for placing the sample to be tested 8 is formed between the first rectangular patch 4 and the second rectangular patch 5.

[0054] The composite circular ring chain 6 is disposed between the first rectangular patch 4 and the second rectangular patch 5, and is located within the sample loading area 7. The composite circular ring chain 6 includes four composite circular ring units 11, which are arranged in a chain along the central region.

[0055] Each composite circular ring unit 11 includes an outermost omega-type resonator structure 12 and two open-ring resonator structures 13 disposed inside the omega-type resonator structure 12. The radii of the two open-ring resonator structures 13 are both smaller than the radius of the omega-type resonator structure 12, and the opening directions of the two open-ring resonator structures 13 are opposite. The open-ring resonator structures 13 are provided with open-ring resonator openings 14, and the open-ring resonator openings 14 and adjacent metal edges form a local capacitive coupling region 15. The omega-type resonator structure 12 is used to concentrate current and microwave magnetic fields, and the open-ring resonator structures 13 and their opening regions are used to enhance the local electric field.

[0056] In this embodiment, the sample loading region 7 is located above the composite circular ring chain 6 and covers part of the first rectangular patch 4, the second rectangular patch 5, and the composite circular ring chain 6. A microfluidic chip is disposed on the sample loading region 7. The microfluidic chip includes a microchannel, a sample inlet, and a sample outlet. The sample inlet is used to communicate with a syringe, allowing the microplastic suspension to flow in from the sample inlet and flow out from the sample outlet after passing through the sample loading region 7 above the composite circular ring chain 6. In a specific simulation example, the orthographic projection of the microchannel on the dielectric substrate corresponds to the sample loading region 7, and the size of the microfluidic chip is 10 mm × 18 mm × 2 mm.

[0057] Because of the electromagnetic coupling between the first rectangular patch 4 and the second rectangular patch 5 and the composite circular chain 6, the composite circular chain 6 can be excited and generate a resonant response when the microwave signal is transmitted between the first port 9 and the second port 10. After the sample under test 8 is loaded onto the sample loading region 7, see... Figure 2 The electromagnetic parameters of the sample, such as dielectric constant and conductivity, will change the equivalent capacitance, equivalent inductance and loss of the sensor, thereby causing resonant frequency shift, amplitude change or phase change in the scattering parameter response.

[0058] This embodiment utilizes finite element electromagnetic simulation software to perform frequency domain simulation of the sensor. During simulation, the sample model to be tested is simplified to a cuboid and placed in the sample loading region 7 between the first rectangular patch 4 and the second rectangular patch 5, above the composite circular chain 6. During the size optimization process, the S values ​​after loading samples of multiple candidate sizes are calculated. 11 The response curves are compared, and the resonant frequency, amplitude attenuation, electric field distribution, and subsequent modeling and fitting requirements are comprehensively analyzed to determine the structural parameters of the sensor.

[0059] In one specific embodiment, the sample model to be tested is loaded onto the sample loading region 7 above the composite circular ring chain 6, and the concentration of the microplastic suspension of the sample is set to 63.60 mg / mL; the simulation frequency range is 2 GHz to 4.5 GHz, and the sampling points are 251. Under this sample loading condition, the sensor exhibits different degrees of resonant response near 2.41 GHz, 2.59 GHz, 2.92 GHz, and 3.56 GHz, respectively. The location of the strongest electric field differs at different frequencies, indicating that the dominant resonant modes of the microstrip transmission line, rectangular patch, and composite circular ring chain in the sensor shift with frequency.

[0060] As shown in Figure 4(a), near 2.41 GHz, the strongest electric field is mainly concentrated in the microstrip transmission line region. Although the composite circular chain is excited, it is not the dominant resonant region. This frequency point mainly exhibits a response driven by the microstrip transmission line resonance. Therefore, this frequency point is at S 11 The amplitude in the curve is relatively shallow, and the sample disturbance effect is relatively weak.

[0061] As shown in Figure 4(b), near 2.59 GHz, S 11 The resonance phenomenon is not obvious on the curve, but a strong electric field distribution appears at the opening of the open-ring resonator on the outer side of the composite circular ring chain and at the edge of the omega-type resonator, especially at the composite circular ring unit at the end of the chain structure. This is because, although the basic dimensions of multiple composite circular ring units are the same, the composite circular ring units at the end of the chain structure have different edge coupling environments, with different equivalent current paths, edge capacitive coupling, and equivalent LC parameters, making it easier to form a strong electric field at the outer edge. Meanwhile, since this strongest electric field mainly occurs at the end and edge regions of the circular ring chain, it has a smaller impact on the main transmission channel formed along the microstrip transmission line and the rectangular patch, and the external coupling at the port is weak. Therefore, at S... 11 The response curve does not show a clear trend.

[0062] As shown in Figure 4(c), near 2.92 GHz, the electric field distribution on the four composite ring units is more uniform, the composite ring chain as a whole is excited, and the coupling between the rings is enhanced. Compared with the end-local mode near 2.59 GHz, this frequency is closer to the overall coupled resonance mode of the composite ring chain. At this frequency, the electric field distribution in the ring chain region is obvious, and the resonant units are more susceptible to changes in the concentration of the sample under test. Therefore, it can be used as an auxiliary detection frequency in this simulation example.

[0063] As shown in Figure 4(d), near 3.56 GHz, S 11 The curve shows the greatest amplitude attenuation, and the electric field is not only distributed along the composite circular ring chain but also significantly enhanced in the rectangular patch and microstrip transmission line regions. This indicates that this frequency point is not a local resonance of a single structure, but rather a strongly coupled resonance mode of the sensor as a whole, formed by the combined effects of the microstrip transmission line, rectangular patch, and composite circular ring chain. Inside the composite circular ring chain, the coupling between adjacent composite circular ring units is enhanced, the coupled modes are jointly excited, the two open-ring resonators resonate simultaneously, and charge accumulation occurs at the openings of the open-ring resonators and adjacent metal edges. This frequency point has a deep S-wave. 11 The amplitude response can be used as the main detection frequency in this simulation example.

[0064] Furthermore, by changing the microplastic concentration of the sample and its corresponding equivalent dielectric parameter, and analyzing the scattering parameter response under different concentration conditions, it was found that as the sample concentration gradually decreases, the main resonant frequency of the sensor shifts generally towards lower frequencies. This is because, as the microplastic concentration in the sample decreases, the equivalent dielectric constant of the sample increases, the equivalent dielectric constant of the sample loading region 7 increases, and the equivalent capacitance of the sensor increases. According to the LC resonance relationship, an increase in equivalent capacitance leads to a decrease in resonant frequency; therefore, S... 11The response exhibits a shift towards lower frequencies. In addition, there are variations on both sides of the resonant frequency band, S... 11 The amplitudes also exhibited monotonic variations. This result indicates that the sensor's resonant frequency and amplitude can produce a recognizable response to changes in the dielectric constant of the microplastic sample, providing a basis for establishing the relationship between microplastic concentration and resonant frequency shift and amplitude.

[0065] To further verify the variation law of sensor scattering parameter response under different microplastic concentrations, this embodiment sets the microplastic suspension concentrations to 0 mg / mL, 17.79 mg / mL, 35.88 mg / mL, 54.28 mg / mL, 73.00 mg / mL and 92.04 mg / mL respectively, and obtains the S11 amplitude, S11 phase and S21 amplitude response curves of the sensor under different concentrations.

[0066] As shown in Figure 3(a), with the increase of microplastic concentration, the main resonant frequency in the S11 amplitude response curve shifts towards higher frequencies overall, while the amplitudes of multiple resonance valleys undergo identifiable changes, consistent with the analysis results obtained above based on the equivalent dielectric constant and LC resonance relationship. Therefore, characteristic parameters such as resonant frequency, fixed-frequency amplitude, and amplitude attenuation can be extracted from the S11 amplitude response.

[0067] As shown in Figure 3(b), the S11 phase response under different microplastic concentrations exhibits significant phase changes near the main resonant frequency band. With increasing microplastic concentration, the phase at a fixed frequency point also shows a distinguishable change; therefore, the phase at a fixed frequency point can be used as a concentration inversion feature.

[0068] As shown in Figure 3(c), the S21 amplitude response curves under different microplastic concentrations exhibit multiple transmission valleys and peaks. The frequencies of the main transmission valleys and the amplitudes at fixed frequency points change regularly with the microplastic concentration. The S21 response can reflect the changes in port transmission and coupling state caused by changes in sample concentration, and can provide complementary characteristic information to the S11 response for microplastic concentration inversion.

[0069] comprehensive Figures 3(a) to 3(c) It is known that different microplastic concentrations cause resolvable changes in the amplitude, phase, and amplitude responses of the sensor's S11 and S21. Since changes in microplastic concentration alter the equivalent dielectric parameters of the sample under test, thereby affecting the sensor's resonant frequency, amplitude, and phase response, multiple sets of microwave characteristic parameters can be extracted from different frequency bands and different scattering parameters, and a correspondence between microplastic concentration and microwave characteristic parameters can be established to obtain the microplastic concentration of the sample under test.

[0070] Specifically, in this embodiment, low-frequency S is established respectively. 11 Model, high frequency S 11 Model and S21 The model has three sub-models. Among them, the low-frequency S... 11 The model uses low-frequency S 11 Resonant frequency, amplitude, and phase are used as input characteristics; high-frequency S 11 The model uses high frequency S 11 Resonant frequency, amplitude, and phase are used as input features; S 21 The model uses S 21 The valley frequency and its corresponding amplitude at a fixed frequency point are used as input features. The expressions for each model are as follows:

[0071]

[0072] Where y1 is the low-frequency S 11 The model predicts the microplastic concentration, y2 is a high-frequency S 11 The model predicts the microplastic concentration, y3 is S 21 Model-predicted microplastic concentration; f L11 For low frequency S 11 Resonant frequency, A L11 For fixed low frequency S 11 Amplitude, P L11 For fixed low frequency S 11 Phase; f H11 For high frequency S 11 Resonant frequency, A H11 To fix S at high frequency 11 Amplitude, P H11 To fix S at high frequency 11 Phase; f F21 For S 21 The valley of the curve corresponds to the frequency, A F21 For S 21 The amplitude corresponding to a fixed frequency point.

[0073] To further improve the inversion accuracy, this embodiment performs a weighted fusion of the three sub-models to obtain the final fused model:

[0074]

[0075] Where y represents the microplastic concentration predicted by the final fusion model. The fitting evaluation metrics for different models are shown in Table 1:

[0076] Table 1

[0077]

[0078] As shown in Table 1, the low-frequency S 11 R of the model 2 The value was 0.99942, MAE was 0.91368, and MARE was 3.3645%; high-frequency S 11 R of the model2 The value was 0.99940, MAE was 1.0439, and MARE was 2.5323%; S 21 R of the model 2 The R-value was 0.99941, MAE was 1.0438, and MARE was 1.9347%; the final fusion model's R-value was... 2 The value is 0.99973, MAE is 0.71453, and MARE is 2.0709%.

[0079] As shown in Figures 5(a), 5(b), and 5(c), the prediction results of the low-frequency S11 model, the high-frequency S11 model, and the S21 model all show good consistency with the actual microplastic concentration, indicating that the microwave features extracted from different frequency bands and different scattering parameters can effectively characterize the changes in microplastic concentration. Compared with the individual sub-models, the prediction results of the fusion model have a higher degree of overlap with the actual concentration, as shown in Figure 5(d). Combined with Table 1, the R² of the fusion model reaches 0.99973, and the MAE decreases to 0.71453 mg / mL, indicating that the multi-feature fusion model has higher fitting accuracy and smaller prediction error than the single-frequency band or single-scattering parameter model, thus improving the accuracy of microplastic concentration inversion.

[0080] like Figure 6 As shown, the absolute errors of the fusion model at different microplastic concentrations are all distributed near the zero error line, with the errors ranging from approximately −1.6 mg / mL to 1.2 mg / mL. Furthermore, the errors do not show a continuous increasing or decreasing trend with increasing microplastic concentration, indicating that the fusion model does not have obvious systematic prediction bias in different concentration ranges and has good prediction stability.

[0081] This embodiment also provides a method for detecting microplastic concentration, which includes the following steps:

[0082] Sample loading. Place the microplastic suspension sample 8 to be tested in the sample loading area 7, so that the sample to be tested covers the sensitive area above the composite circular chain 6.

[0083] Microwave frequency sweep. Connect the first port 9 and the second port 10 to a vector network analyzer or microwave signal acquisition module to sweep the frequency of the sensor in the range of 2 GHz to 4.5 GHz, and acquire the sensor's S... 11 S 21 Response and corresponding phase response.

[0084] Feature extraction. Microwave feature parameters are extracted from the scattering parameter response, including resonant frequency, resonant frequency offset, amplitude attenuation, phase magnitude, and combined features between different resonant frequencies.

[0085] Model establishment. A correspondence between microplastic concentration and microwave characteristic parameters is established using samples with known microplastic concentrations. In a specific implementation, the resonant frequency shift, amplitude attenuation, and phase magnitude of the main detection frequency are used as primary features, while the resonant frequency shift, amplitude attenuation, and phase magnitude of the auxiliary detection frequency are used as auxiliary features. A prediction model is established through multi-parameter fitting or machine learning.

[0086] Concentration detection. For test samples with unknown microplastic concentrations, microwave characteristic parameters are extracted, and the microplastic concentration detection results are obtained based on an established prediction model.

[0087] In summary, this invention forms a localized, highly sensitive electric field within the sample loading region using a composite circular chain structure, and achieves high-precision detection of microplastic concentration in water based on multidimensional microwave characteristic parameters under multi-resonance response.

Claims

1. A chain-type composite multi-ring resonant microwave planar microplastic concentration detection sensor, characterized in that, include: Dielectric substrate; The first microstrip transmission line and the second microstrip transmission line are disposed on the upper surface of the dielectric substrate; A first rectangular patch and a second rectangular patch are connected and coupled to the first microstrip transmission line and the second microstrip transmission line. A sample loading region is formed between the first rectangular patch and the second rectangular patch. as well as A composite circular ring chain is disposed between the first rectangular patch and the second rectangular patch and located within the sample loading area; the composite circular ring chain includes multiple composite circular ring units arranged in a chain along the central axis, each of the composite circular ring units including an outermost omega-type resonator structure and two open-ring resonator structures disposed inside the omega-type resonator structure; The radii of the two concentric open-ring resonator structures are both smaller than the radius of the omega-type resonator structure, and the opening directions of the two open-ring resonator structures are opposite. The omega-type resonator structure is used to concentrate surface current and microwave magnetic field. The openings of the two open-ring resonator structures and adjacent metal edges form a local capacitive coupling region, which is used to form a local strong electric field in the sample loading region.

2. The sensor according to claim 1, characterized in that, The number of composite ring units is four, and the four composite ring units are arranged in a chain along the central region.

3. The sensor according to claim 1 or 2, characterized in that, The sample loading area is located above the composite circular chain and covers part of the first rectangular patch, the second rectangular patch, and the composite circular chain.

4. The sensor according to claim 1, characterized in that, The sensor adopts a dual-port structure. The side of the dielectric substrate is provided with a first port and a second port. The first port is connected to the first microstrip transmission line, and the second port is connected to the second microstrip transmission line. The first port and the second port are used to input and output microwave excitation signals.

5. The sensor according to claim 4, characterized in that, The first port and the second port are connected to a vector network analyzer or a microwave signal acquisition module.

6. The sensor according to claim 1, characterized in that, A microfluidic chip is disposed on the sample loading area, and the microfluidic chip includes a microchannel, a sample inlet, and a sample outlet; The sample inlet is used to communicate with the syringe, allowing the microplastic suspension to flow in through the sample inlet and out through the sample outlet after passing through the sample loading area above the composite circular chain.

7. A method for detecting microplastic concentration using the chain-type composite multi-ring microwave microplastic concentration detection sensor according to any one of claims 1 to 6, characterized in that, Includes the following steps: (1) Place the sample to be tested in the sample loading area so that the sample to be tested is above the composite circular chain; (2) The scattering parameter response of the sensor is acquired within a preset frequency range using a vector network analyzer or a microwave signal acquisition module; (3) Extract the resonant frequency, amplitude attenuation and phase magnitude of multiple resonant frequency points from the scattering parameter response as microwave characteristic parameters; (4) Establish the correspondence between microplastic concentration and microwave characteristic parameters based on known microplastic concentration samples; (5) Test the sample with unknown microplastic concentration and obtain the microplastic concentration test result according to the corresponding relationship.

8. The method according to claim 7, characterized in that, The specific steps of acquiring the scattering parameter response of the sensor within the preset frequency range are: The sensor was frequency-scanned within the range of 2 GHz to 4.5 GHz, and the S-band of the sensor was collected. 11 S 21 Response and corresponding phase response.

9. The method according to claim 7 or 8, characterized in that, Extracting microwave characteristic parameters from the scattering parameter response also includes extracting the resonant frequency offset and the combined features between different resonant frequency points.

10. The method according to claim 9, characterized in that, The process of establishing the correspondence between microplastic concentration and microwave characteristic parameters includes: using the resonant frequency shift, amplitude attenuation, and phase magnitude of the main detection frequency as the main features, and the resonant frequency shift, amplitude attenuation, and phase magnitude of the auxiliary detection frequency as auxiliary features, and performing multi-parameter fitting or machine learning to establish a prediction model.

Images